/WS client & ABORTED transactions

Hello all,
I am bit confused with "Native client" concept.
What is the diffrence between a native client & a remote (/WS ) client?
As far as i know,remote clients are those which are connected to the front end application via workstations.
I have a huge billing application residing over tux(7.1).
what is diffrence between application client & remote tux client? is it the same?
i use following command to check the users connected to tuxedo.
echo 'pclt' | tmadmin -r 2>/dev/null | egrep -v '>|-------|tpsysadm|tmadmin|WSH|^$'|wc -l
sometimes i see around 60 users connected...& i have configured only 50 WS clients in my ubb file.
is there any way i can remove unwanted users ?
also i need to clean up transactions which are in ABORTED state?
Can't we control users & transactions via tux commands?
regards
abhijeet

Workstation client is the client connected to Tuxedo with WSL/WSH. Native
client accesses the IPC queue directly. The pclt of tmadmin is different for
native client and workstation client. The later has a "/W" flag in the
status cloumn.
You can define the MAXACCESSERS in ubbconfig file to limit the users. For
example,
MAXACCESSERS = 100
MAXSERVERS=20
MAXWSCLIENTS=70
Then you leave 10 slots for WSH and tmadmin etc.
Wayne
<Abhijeet K> wrote in message news:[email protected]..
Hello all,
I am bit confused with "Native client" concept.
What is the diffrence between a native client & a remote (/WS ) client?
As far as i know,remote clients are those which are connected to the frontend application via workstations.
>
I have a huge billing application residing over tux(7.1).
what is diffrence between application client & remote tux client? is itthe same?
>
i use following command to check the users connected to tuxedo.
echo 'pclt' | tmadmin -r 2>/dev/null | egrep -v'>|-------|tpsysadm|tmadmin|WSH|^$'|wc -l
>
sometimes i see around 60 users connected...& i have configured only 50 WSclients in my ubb file.
>
is there any way i can remove unwanted users ?
also i need to clean up transactions which are in ABORTED state?
Can't we control users & transactions via tux commands?
regards
abhijeet

Similar Messages

Transient persestant memory-begin abort transaction-arraycopynonatomic

If array defined as transient, then abort doesnt make any sense, chages are not rolled back.(Both for arrayCopy and arraycopyNonAtomic functions)
     public void AtomicNonatomic()
byte hello[] = {'H','E','L','L','O'};
          byte[] key_buffer = JCSystem.makeTransientByteArray((byte)16, JCSystem.CLEAR_ON_RESET);
          //byte[] key_buffer = new byte[16];
          JCSystem.beginTransaction();
          //Util.arrayCopy(hello, (short)0, key_buffer, (short)0, (short)5);
          Util.arrayCopyNonAtomic(hello,(short)0, key_buffer, (short)0, (short)5);
          for(byte i =0; i<2; i++)
          key_buffer=0; //This types of assigning values are atomic operations, so can be thougth as arrayCopy
          JCSystem.abortTransaction();
If array not defined as transient and if arrayCopy is used then abort works
If array not defined as transient and if arrayCopyNonatomic is used then abort not works
     public void AtomicNonatomic()
          //byte[] key_buffer = JCSystem.makeTransientByteArray((byte)16, JCSystem.CLEAR_ON_RESET);
          byte[] key_buffer = new byte[16];
          JCSystem.beginTransaction();
          //Util.arrayCopy(hello, (short)0, key_buffer, (short)0, (short)5);
          Util.arrayCopyNonAtomic(hello,(short)0, key_buffer, (short)0, (short)5);
          for(byte i =0; i<2; i++)
          key_buffer[i]=0;
          JCSystem.abortTransaction();
As a result begin-abort-commit should be used with non-transient arrays and without arrayCopyNonatomic function.
Edited by: simeranya_00 on Aug 7, 2010 6:51 AM
Edited by: simeranya_00 on Aug 7, 2010 6:56 AM

Check out JCRE specification. Therein you will find what is (not) part of a transaction.
1. All global arrays are not part of a transaction
2. All transient fields/arrays/keys are not part of a transaction
3. OwnerPIN object is not part of a transaction
4. All Java Card API methods are transactional, except arrayCopyNonAtomic and arrayFillNonAtomic
Lifetime of a transaction
begin <--> commit
begin <--> abort < -- rollback
begin <--> JCRE context (process method exit) < -- abort transaction (done by JCRE) < -- rollback
begin <--> uncaught exception < -- abort < -- rollback
begin <--> tear/reset < -- rollback

Errors ajp_process_callback::jk_ajp_common.c (1954): Writing to client aborted or client network

Hello,
we have the following configuration:
S.O. : Red Hat Enterprise Linux Server release 6.5
Apache : 2.2.15
Coldfusion 10 Standard update 12
SElinux disable
Iptables disable
Configuring / etc / httpd / conf / mod_jk.conf
# Load mod_jk module
Jk_module LoadModule " / opt/coldfusion10/config/wsconfig/1/mod_jk.so "
# Where to find workers.properties
JkWorkersFile " / opt/coldfusion10/config/wsconfig/1/workers.properties "
JkMountFile " / opt/coldfusion10/config/wsconfig/1/uriworkermap.properties "
# Where to put jk logs
JkLogFile " / opt/coldfusion10/config/wsconfig/1/mod_jk.log "
# Where to put jk shared memory
JkShmFile " / opt/coldfusion10/config/wsconfig/1/jk_shm "
# Set the jk log level [debug / error / info]
JkLogLevel info
# Select the timestamp log format
JkLogStampFormat " [ % a% b % d% H:% M: % S % Y ]"
AddHandler jakarta- servlet. Cfm . Cfml . Cfc . Cfr. Cfswf
In the log / opt/coldfusion10/config/wsconfig/1/mod_jk.log observe the following errors:
[ Tue Dec 03 10:29:53 2013] [ 3754:140076806252512 ] [info ] jk_ajp_common.c ajp_process_callback :: (1954 ) : Writing to client aborted or client network problems
[ Tue Dec 03 10:29:53 2013] [ 3754:140076806252512 ] [info ] jk_ajp_common.c ajp_service :: ( 2629 ): ( cfusion ) sending request to tomcat failed ( unrecoverable ) , Because of client write error ( attempt = 1 )
[ Tue Dec 03 10:29:53 2013] [ 3754:140076806252512 ] [info ] jk_handler :: mod_jk.c ( 2718 ) : Aborting connection for worker = cfusion
Can anyone tell how to solve this problem because the connection between apache and coldfusion degrade coming to collapse the system.

Not my particular environment Linux Apache however the way I read the mod_jk log content is very similar or same as this other thread and detail in isapi_redirect log on Windows IIS environment.
Thread:
http://forums.adobe.com/message/5878493#5878493
MOD_JK log sample:
[info ] jk_ajp_common.c ajp_process_callback :: (1954 ) :
Writing to client aborted or client network problems
[info ] jk_ajp_common.c ajp_service :: ( 2629 ):
( cfusion ) sending request to tomcat failed ( unrecoverable ) ,
Because of client write error ( attempt = 1 )
[info ] jk_handler :: mod_jk.c ( 2718 ) :
Aborting connection for worker = cfusion
ISAPI_redirect log sample:
[info] ajp_process_callback::jk_ajp_common.c (1984):
Writing to client aborted or client network problems
[info] ajp_service::jk_ajp_common.c (2684):
(cfusion) sending request to tomcat failed (unrecoverable),
because of client write error (attempt=1)
[info] HttpExtensionProc::jk_isapi_plugin.c (2289):
service() failed because client aborted connection
Where I am seeing that sequence of log, CF10 IIS is not having any issue so it may not actually relate to the CF10 not responding. I would like to find a glossary of tomcat messages and the meaning however I have not got an
interpretation to that log sequence. I will add the sites for me that have those logs have BYO wireless clients and it would be fair to say those users do not log off CF application however just close their browser or lid of laptop. Neither have I been able to reproduce the environment to cause that log detail so merely guess from me rather than statement of fact.
HTH, Carl.

"Transaction Errors : Aborting transaction on session " while processing the cube

HI Team,
Currently i have developed a cube and successfully deployed it in to the SSAS server.
But when i process the cube the measures in the cube got successfully processed. After that the process is still running and showing the status as "Transaction errors : aborting transaction on session XYZAB".
can you please guide me in solving this issue. The cube takes more than 6 hrs to process.
thanks in advance
baskar k

Hi,
I have Similar issue with 2005 and in 2005 I can't execute select * from $system.discover_sessions.
Do we have any other way to resolve it.
If I restart SSAS Server, It starts working fine and I cant restart at day time.
http://blogs.msdn.com/b/sql_pfe_blog/archive/2009/08/27/deadlock-troubleshooting-in-sql-server-analysis-services-ssas.aspx
Thanks Shiven:) If Answer is Helpful, Please Vote

Why is a JSP page or servlet (SingleThreadModel) unreachable on iPlanet 6.0 webserver after client abort?

After client abort (user pushes STOP on the browser, etc.) JSP page or servlet (SingleThreadModel) is unreachable on iPlanet 6.0 webserver. Only a server restart helps. Can anybody help me?

adsfadsfaf

Client Extension: Transaction Cost

Hello
We have implemented Account Generation which allow us to provide alternative accounts regarding our requirements.
Now I have bigger task: I need change transaction value (or item cost) according to source subinventory. Exact I have part of items where my bookkeepers don't want any distribution or they allow distribution with '0' value. I thought that the better way is Client Extension: Transaction Cost Extension. Unfortunately the documentation is very poor or I cannot find it.
If anybody have any examples or a good documentation I will be appreciate for any help.
Anyway, my knowledge at this time is following:
I need customize a function "actual_cost_hook" in CSTPACHK but I don't know how.
A few words about my instance:
This is eBS 11.5.10.2 on RH4.0. We've worked in FIFO organization in multiorganization mode.
Thanks advance
Krzysztof

Hi,
It's very easy, you can call make use this actual_cost_hook function which is going to called from Cost manager package CSTPAVCP.
If you make this function returns for specific transactions as 0 value, your job done.
Regards,
Sarvottam.

Client Deletion, Transaction log getting full.

Hi Gurus,
We are trying to delete a client by running:
clientremove
client = 200 (*200 being the client we want to remove)
select *
The transaction log disk space allocated is 50GB, it is getting full (in simple mode) and client deletion never completes. The size of the table it is accessing is 86 GB, and i think 200 client will be occupying around 40-45GB. Client 200 has 15.5 million rows in the table.
I am i giving proper command ?is there any explicit commit i can include or any workaround for deleting the client and not hammer the log file.
Thanks guys
Edited by: SAP_SQLDBA on Jan 22, 2010 6:51 PM

Hi,
Backup the active transaction log file and Shrink the file directly.
Please refer the following SAP Notes to get more information.
[ Note 625546 - Size of transaction log file is too big|https://websmp130.sap-ag.de/sap%28bD1lbiZjPTAwMQ==%29/bc/bsp/spn/sapnotes/index2.htm?numm=625546]
[ Note 421644 - SQL error 9002: The transaction log is full|https://websmp130.sap-ag.de/sap%28bD1lbiZjPTAwMQ==%29/bc/bsp/spn/sapnotes/index2.htm?numm=421644]
Which version of SQL Server u are using ? SP Level ?
Frequently perform Transaction Log backup (BACKUP TRANS) to remove inactive space within the Transaction Log Files.
Please refer [Note 307911 - Transaction Log Filling Up in SQL Server 7.0|https://websmp130.sap-ag.de/sap%28bD1lbiZjPTAwMQ==%29/bc/bsp/spn/sapnotes/index2.htm?numm=307911] to get more information about the reasons for such kind of situation.
Regards,
Bhavik G. Shroff

Client side transaction support?

Hi,
I am curious if anyone knows of any client side AS3/flex
transaction library out there. Specifically client side, not
anything server side.
For example I have one operation that does 2 distinct
operations, if any one of them fails I need to rollback both of
them. An example would be a class that is responsible for modifying
a file on disk and then updating a local database on the client
(regardless of implementation sqllite, xml, textfile etc). Should
either of these operations fail I would like to rollback the
file/db to it's original state.
Does anyone know of any client side AS3 transaction library
that defines interfaces for such a framework? Thanks!

The problem is that NTLM is a Microsoft proprietary authentication mechanism.
I believe there are some products in the JavaMail Third Party Products page
that support NTLM.
http://java.sun.com/products/javamail/Third_Party.html

Client Abort Exception

Hi all ,
I have a chaching servlet and a compression filter servlet. the compression filter servlet througs up ClientAbortException . what could be the reason.
I know that this exception is thrown when the user closes the browser when the responses is sent of stop the action through browser. but in my case nothing is the case.
Regards,
krthekeyan.s

Well you need to understand the behavior of h-reap or what it's called now, FlexConnect. In this mode, the clients are still remembers on the WLC until the session timer/idle timer expires. So switching between SSID's in h-reap will not be the same when switching when the AP's are in local mode.
Take a look at the client when connected in FlexConnect in the WLC GUI monitor tab. Thus will show you what ssid and vlan the client is on. Now switch to a different ssid and compare this. It's probably the same because the client has not timed out. Now go back to the other ssid and look again. Now on the WLC, remove or delete the client and then switch to the other ssid at the same time. Or switch SSID's and then remove the client. The client will join the new ssid and in the monitor tab, you should see the info.
There is no need to have clients have multiple SSID's unless your testing. Devices should only have one ssid profile configured to eliminate any connectivity issues from the device wanting to switch SSID's.
Sent from Cisco Technical Support iPhone App

Local client copy (Transaction SCCL)

Hi Experts
I am a Basis tem member I have installed SAP Netweaver 2004s now I am create (Tr. Code SCCL) a new client as 100 when I am login in 100 client as sap* and password is pass it is not logged
Can u help me how to create new client and how to copy process
Please help me
With best regards
Sanjay C.

Hi,
When you create a new client from SCC4, you can login to this newly created client using user - SAP* and password - PASS.
If you are not able to login using this credentials then check the parameter login/no_automatic_user_sapstar (as Juan mentioned above).
The user SAP* is inactive by default in a new client. To activate the user SAP*, set the parameter login/no_automatic_user_sapstar = 0, and restart the system.
You should be able to login after this.
For local client copy, you can check the link - http://help.sap.com/saphelp_nw04/helpdata/en/69/c24c5b4ba111d189750000e8322d00/frameset.htm
for details.
Its very simple, you need to ensure that enough space is there, select the required profile and start the copy in SCCL.
You can monitor the same via SCC3.

Client demarkation of transaction and multiple clients

          Hi!
          I am using client demarkated transaction on stateless session bean
          in WLServer6.0. The client runs outside of WLServer.
          How can I get two different UserTransaction objects in two different
          clients? What if, one client tries to use two EJBs in the same
          app-server?
          If I use following code to get the UserTransaction object, transaction
          of two clients get mixed up and then messes the functionality.
          -- Code starts --
          Properties env = new Properties();
          env.put(Context.INITIAL_CONTEXT_FACTORY," weblogic.jndi.WLInitialContextFactory");
          env.put(Context.PROVIDER_URL, "t3://AppServer:7001");
          Context initial = new InitialContext(env);
          UserTransaction ut = (UserTransaction) initial.lookup ("javax.transaction.UserTransaction");
          ut.begin (); ....... // call ejb beans on AppServer ....... ut.commit


Vinay wrote:
          >
          > Hi!
          > I am using client demarkated transaction on stateless session bean
          > in WLServer6.0. The client runs outside of WLServer.
          > How can I get two different UserTransaction objects in two different
          > clients?
          I don't understand the question. UserTransaction is a bit mis-named.
          It allows you to start, commit, and abort transactions, but it is not
          itself a transaction.
          > What if, one client tries to use two EJBs in the same
          > app-server?
          >
          That's fine.
          > If I use following code to get the UserTransaction object, transaction
          > of two clients get mixed up and then messes the functionality.
          >
          You'll have to provide more information. Your code looks fine.
          -- Rob
          > -- Code starts --
          >
          > Properties env = new Properties();
          > env.put(Context.INITIAL_CONTEXT_FACTORY," weblogic.jndi.WLInitialContextFactory");
          >
          > env.put(Context.PROVIDER_URL, "t3://AppServer:7001");
          > Context initial = new InitialContext(env);
          > UserTransaction ut = (UserTransaction) initial.lookup ("javax.transaction.UserTransaction");
          >
          > ut.begin (); ....... // call ejb beans on AppServer ....... ut.commit
          > ();
          Coming Soon: Building J2EE Applications & BEA WebLogic Server
          by Michael Girdley, Rob Woollen, and Sandra Emerson
          http://learnweblogic.com

Abort processing when error encountered in CALL TRANSACTION

Is there a way to abort transaction in CALL TRANSACTION <tcode> USING <bdcdata> when an error has encountered?
If there is, what should it be?

In CALL TRANSACTION the standard behavior of the called transaction will be replicated. So if in VA01 the billing plan population is aborted if an error is encountered your CALL TRANSACTION will behave the same way.
BR,
Suhas

Get IllegalStateException when rollback client transaction while thansaction is already rollbacked by Container

Hi,
          I use WL5.1.0 + SP8. When I start a client-demarcated transaction and
          call two methods in EJB, the second method will throw a system level
          exception (such as RemoteException), the container rollback the
          transaction so that when I catch the exception in client and try to
          rollback by myself, I got an IllegalStateException: There is no
          transaction associated with this thread.
          In SP8, I thought they fixed this problem and the container is supposed
          to mark the transaction rollbacked and throw
          TransactionRolledbackException to client instead of rollback this
          transaction.. But obviously it still doesn't work in SP8.
          Is this still a bug or am I missing something?
          Thanks a lot!
          Tao


Hi,
          I use WL5.1.0 + SP8. When I start a client-demarcated transaction and
          call two methods in EJB, the second method will throw a system level
          exception (such as RemoteException), the container rollback the
          transaction so that when I catch the exception in client and try to
          rollback by myself, I got an IllegalStateException: There is no
          transaction associated with this thread.
          In SP8, I thought they fixed this problem and the container is supposed
          to mark the transaction rollbacked and throw
          TransactionRolledbackException to client instead of rollback this
          transaction.. But obviously it still doesn't work in SP8.
          Is this still a bug or am I missing something?
          Thanks a lot!
          Tao

NPE on Transaction.abort()

Hello. I am using the latest version of je 3.3.75.
I am getting a NPE when I believe abort() is called twice:
Caused by: java.lang.NullPointerException
     at com.sleepycat.je.Transaction.abort(Transaction.java:90)
In the code, here's where I believe this is coming from:
90:      txn.abort(false); // no sync required
This seems like a really small issue. It would be nice to be able to tell the state of a transaction. At the very least, an error besides NPE would be necessary.
Where is the issue tracker for sleepycat?
Cheers,
-Adrian

I'm sorry Adrian, I didn't answer your question earlier.
While BDB is available under an open-source license, we don't make public some of the facilities that other projects do. All of the following are private to Oracle and will remain so for the foreseeable future:
- Issue tracking system
- Source control system
- Road map and other internal documentation
The ticket number for this fix is #16214, and you'll see that in the change log when the release with the fix comes out.
Thanks,
--mark

Aborted multi-machine transaction stays in BB

I have a distributed transaction going across 2 machines. I'm testing
out what happens when I abort the transaction. If I go into tmadmin,
and do a pt, I get the following printout. The problem is the
OK_ORATMSGRP in the first machine is specified being remote and as the
coord group and has a status of TMGACTIVE. The same group on the other
machine (where the group is local) has a status of TMGABORTED. Now this
is exactly the same group, but one machine thinks it is TMGACTIVE and
the other machine thinks it's TMGABORTED (this is the correct status).
The transaction will remain in the bulletin board until the timeout
value expires. This shouldn't happen. I thought as soon as I called
tx_rollback the whole transaction should be aborted. What is going on?
This is causing a problem because before we didn't specify a transaction
timeout and the BB was being filled with Aborted transactions that had
that one group marked as TMGACTIVE when it shouldn't have been.
Subsequently the BB became full and new transactions couldn't be
started.
index=2 gtrid=x0 x3d25495d xa61
: Machine id: IRGVELOCITY, Transaction status: TMGABORTED
Group count: 2, timeout: 30, transaction timed out
Known participants:
group: OK_ORATMSGRP, status: TMGACTIVE, remote, coord
group: OK_FXNTGRP, status: TMGABORTED, local
Machine id: CDDEV, Transaction status: TMGABORTED
Group count: 4, timeout: 30, transaction timed out
Known participants:
group: OK_ORATMSGRP, status: TMGABORTED, local, coord
group: OK_MQTMSGRP, status: TMGABORTED, local
group: OK_TUXTMSGRP, status: TMGABORTED, local
group: OK_FXNTGRP, status: TMGABORTED, remote

So it would be fair to say that as a rule, you should always specify a reasonably
low transaction timeout so the BB doesn't become full of aborted transactions? I
still think that an aborted transaction should be removed from the BB at some point
(especially for aborted transactions with no timeout value). Maybe there should
be a different timeout value, separate from the transaction timeout value that determines
how long an aborted transaction should remain in the bulletin board.
Peter Holditch <[email protected]> wrote:
Frank,
The timeout is in there to prevent the scenario where a transaction aborts
and
is deleted, and then another message on behalf of the same transaction arrives.
With the records having been deleted, the system has no option but to
create a
new GTT entry for the apparently new transaction and then wait for it to
time out.
I know that this timeout is also in the WLS implementation. I do not believe
that it is infinite!
Note also that the only trandactions that will ever hit the TLOG are ones
that
have decided to commit, by which time this timeout is not relevant any more.
That's my recollection of the reasoning, I hope it sheds some light...
Regards,
Peter.
Frank Clarijs wrote:
Failed (rolled back) distributed transactions remain in the BB until
timeout, by design. I'm not sure why, but this behaviour is the one BEA
(or its predecessors) explicitely implemented.
This has of course strange consequences: if your application aborts
transactions quite often, either you need a big TLOG and even higher
setting of "maximum global transactions" in the UBB (or you can't
restart your application after a crash), or you need to set the timeout
quite low, with all risks from that...
Don't worry, it's even worse in Weblogic, where the bean for an aborted
transaction never disappears, by design.
Frank
Anthony Fryer wrote:
I have a distributed transaction going across 2 machines. I'm testing
out what happens when I abort the transaction. If I go into tmadmin,
and do a pt, I get the following printout. The problem is the
OK_ORATMSGRP in the first machine is specified being remote and as the
coord group and has a status of TMGACTIVE. The same group on the other
machine (where the group is local) has a status of TMGABORTED. Now
this is exactly the same group, but one machine thinks it is TMGACTIVE
and the other machine thinks it's TMGABORTED (this is the correct
status). The transaction will remain in the bulletin board until the
timeout value expires. This shouldn't happen. I thought as soon as I
called tx_rollback the whole transaction should be aborted. What is
going on? This is causing a problem because before we didn't specifya
transaction timeout and the BB was being filled with Aborted
transactions that had that one group marked as TMGACTIVE when it
shouldn't have been. Subsequently the BB became full and new
transactions couldn't be started.
index=2 gtrid=x0 x3d25495d xa61
: Machine id: IRGVELOCITY, Transaction status: TMGABORTED
Group count: 2, timeout: 30, transaction timed out
Known participants:
group: OK_ORATMSGRP, status: TMGACTIVE, remote, coord
group: OK_FXNTGRP, status: TMGABORTED, local
Machine id: CDDEV, Transaction status: TMGABORTED
Group count: 4, timeout: 30, transaction timed out
Known participants:
group: OK_ORATMSGRP, status: TMGABORTED, local, coord
group: OK_MQTMSGRP, status: TMGABORTED, local
group: OK_TUXTMSGRP, status: TMGABORTED, local
group: OK_FXNTGRP, status: TMGABORTED, remote

/WS client & ABORTED transactions

Similar Messages

Maybe you are looking for